Currently, there is limited research on entity relation extraction in the domain of ancient Chinese, primarily due to the scarcity of publicly available datasets for model training. To advance the study of entity relation extraction tasks in ancient Chinese, this paper reconstructs a CMAG-ERE2.0 dataset based on "Comprehensive Mirror to Aid in Government" Addressing issues such as missing relation predictions and inaccurate long-span entity boundary recognition prevalent in existing models applied to ancient Chinese texts, we propose a joint entity relation extraction model tailored for this context. In our model, relative position information of tokens enhances constraints on long-span entity boundaries, while multi-step fusion of relationship type information establishes connections between relational hints and entities, facilitating precise extraction of both long-span entities and latent relations. Additionally, we implement multi-task decoding for modeling purposes, which significantly improves the model’s ability to extract entities and relations within complex textual environments. Experimental results demonstrate that our proposed model effectively mitigates these challenges.

Programming Knowledge Tracing (PKT) aims to estimate students' programming proficiency by analyzing their historical coding activities. It is important to leverage interactive information for PKT code representations, as code submissions are problem-specific interactive solutions evaluated by Online Judge (OJ) feedback. However, most existing PKT models only rely on pre-trained language models to capture static code information, overlooking feedback scores and problem-specific context. To address this issue, we propose an Interactive Information integrated Code Embedding for Programming Knowledge Tracing (IICE-PKT). This model learns a unified code representation by integrating comprehensive interactive information, including code text, OJ feedback scores, problem statement text, problem-skill correlations, and problem difficulties. IICE-PKT consists of three modules: Problem Representation module generates enhanced problem embeddings based on problem-skill correlations and problem difficulties; Code Representation module combines fine-tuned CodeBERT-based code text embeddings, GPT-generated problem text embeddings, and enhanced problem embeddings with supervised feedback scores to produce a unified code representation; Dual-Sequence Modeling module employs two independent GRUs to separately model the problem sequence and the code sequence. Extensive experiments demonstrate the superiority of IICE-PKT. Ablation studies confirm that integrating interactive information into code representations significantly enhances effectiveness.

ABSTRACT The Diffusion Probabilistic Model (DM) has emerged as a powerful generative model in the field of image synthesis, capable of producing high‐quality and realistic images. However, training DM requires a large and diverse dataset, which can be challenging to obtain. This limitation weakens the model's generalisation and robustness when training data is limited. To address this issue, EDG‐CDM, an innovative encoder‐guided conditional diffusion model was proposed for image synthesis with limited data. Firstly, the authors pre‐train the encoder by introducing noise to capture the distribution of image features and generate the condition vector through contrastive learning and KL divergence. Next, the encoder undergoes further training with classification to integrate image class information, providing more favourable and versatile conditions for the diffusion model. Subsequently, the encoder is connected to the diffusion model, which is trained using all available data with encoder‐provided conditions. Finally, the authors evaluate EDG‐CDM on various public datasets with limited data, conducting extensive experiments and comparing our results with state‐of‐the‐art methods using metrics such as Fréchet Inception Distance and Inception Score. Our experiments demonstrate that EDG‐CDM outperforms existing models by consistently achieving the lowest FID scores and the highest IS scores, highlighting its effectiveness in generating high‐quality and diverse images with limited training data. These results underscore the significance of EDG‐CDM in advancing image synthesis techniques under data‐constrained scenarios.

Visual storytelling (VIST) involves generating coherent, creative, and vivid narrative for a collection of images. It remains an immense challenge within cross-modal domain. Traditional mainstream storytelling work were less proficient in handling long sequential relationships. Though large-scale visual-language pre-training (VLP) models demonstrated promising prospect on cross-modal tasks. So far they still were not particularly adept at handling tasks involving image sequences. Moreover, the reference descriptions in the available VIST benchmark dataset are short and simplistic, which constrains model’s potential capabilities. Current models struggle to produce truly rich and vivid narratives. Therefore, in this article, we will address these deficiencies, and contribute from both dataset and innovative model aspects. Firstly, by leveraging large language model (LLM), we have constructed a new dataset VIST++ which can enrich vivid narratives for open-ended image sequences. The dataset has potential on providing beneficial support for future model learning. Secondly, we have proposed an innovative auto-regressive story generation model named ReStoryGen. It can be applied to image sequences of varying lengths in an open-ended way. We have performed extensive experiments and evaluations in terms of visual grounding, coherence and non-redundancy. The experiment results have convincingly demonstrated ReStoryGen achieves impressive outcomes through utilizing parameter-efficient instruction-tuning. Related source codes and models are distributed on Github https://github.com/lixinliu1995/story_gen .

Dance is a creative performance art and must keep coherent with the rhythm and style of music. To address these issues, most of the existing music-driven dance synthesis methods utilize deep generative models and capture the dynamic characteristics of dance motions. However, we observe that dance motions contain big-scale body part movements and small-scale joint movements that are mutually coordinated and related, and the generated dance motions, particularly affected by$L_{1}$loss, are too restrictive and conservative. In this paper, we propose a multi-scale cascaded music-driven dance synthesis network (MC-MDSN) that first generates big-scale body motions conditioned on music and then further refines local small-scale joint motions. Furthermore, we design a multi-scale feature loss to capture the dynamic characteristics of each scale motions and the relations between different scale motion joints. Experimental results show that our method generates better dance motions than the baselines.

It is necessary for the music-to-dance generation to consider both the kinematics in dance that is highly complex and non-linear and the connection between music and dance movement that is far from deterministic. Existing approaches attempt to address the limited creativity problem, but it is still a very challenging task. First, it is a long-term sequence-to-sequence task. Second, it is noisy in the extracted motion keypoints. Last, there exist local and global dependencies in the music sequence and the dance motion sequence. To address these issues, we propose a novel autoregressive generative framework that predicts future motions based on past motions and music. This framework contains a music conformer, a motion conformer, and a cross-modal conformer, which utilizes the conformer to encode music and motion sequences, and further adapt the cross-modal conformer to the noisy dance motion data that enable it to not only capture local and global dependencies among the sequences but also reduce the effect of noisy data. Quantitative and qualitative experimental results on the publicly available music-to-dance dataset demonstrate our method improves greatly upon the baselines and can generate long-term coherent dance motions well-coordinated with the music.

Automatic image colorization without manual interventions is an ill-conditioned and inherently ambiguous problem. Most of existing methods focus on formulating colorization as a regression problem and learn parametric mappings from grayscale to color through deep neural networks. Due to the multimodalities of color-grayscale space, in many applications, it is not required to recover exact ground-truth color. Pair-wise pixel-to-pixel learning-based algorithms lack rationality. Techniques such as color space conversion techniques are then proposed to avoid such direct pixel learning. However, the coloring results after color space conversion are blunt and unnatural. In this paper, we hold viewpoints that a reasonable solution is to generate some colorized result that looks natural. No matter what color a region is to be assigned, the colorized region should be semantically and spatially consistent. In this paper, we propose an effective semantic-aware automatic colorization model via unpaired cycle-consistent self-supervised network. Low-level monochrome loss, perceptual identity loss and high-level semantic-consistence loss, together with adversarial loss, are introduced to guide network self-training. We train and test our model on randomly selected subsets from PASCAL VOC 2012. The experimental results including human subjective studies demonstrate that, compared with state-of-the-art methods, our proposed model can achieve more convincing and superior results. Relevant source code is available at https://github.com/YuSuen/ACCycleGAN.

As an effective global optimization technique, artificial bee colony (ABC) algorithm has become one of the hottest research topics in the fields of evolutionary algorithms. However, the solution search equation is not rotationally invariant, which causes the problem that the performance of ABC is sensitive to the coordinate system. Although many improved ABC variants have been developed, they rarely considered the problem. Hence, to solve the problem, we propose a new ABC variant with bi-coordinate systems (BSABC), including the original coordinate system and the eigen coordinate system. The two coordinate systems own different characteristics: (1) the former one aims to maintain the population diversity, and (2) the latter one is to adapt the search to the fitness landscape of the problems. Based on the characteristics, in the BSABC, the two coordinate systems are used in the employed bee phase and onlooker bee phase, respectively. Meanwhile, two new solution search equations are designed by utilizing the elite information, and they are respectively performed in the two coordinate systems to further improve the algorithm performance. As another contribution of this study, in the scout bee phase, the multivariate Gaussian distribution is constructed to replace the original method to generate offspring, which is helpful to save the search experience. The performance of the BSABC is verified by extensive experiments on the CEC2013 test suite and one real-world optimization problem, and four well-established ABC variants and three other evolutionary algorithms are included in the performance comparison. The comparison results confirm that the BSABC shows competitive performance by achieving better results on the majority of test functions.

Chinese dialects discrimination is a challenging natural language processing task due to scarce annotation resource. In this article, we develop a novel Chinese dialects discrimination framework with transfer learning and data augmentation (CDDTLDA) in order to overcome the shortage of resources. To be more specific, we first use a relatively larger Chinese dialects corpus to train a source-side automatic speech recognition (ASR) model. Then, we adopt a simple but effective data augmentation method (i.e., speed, pitch, and noise disturbance) to augment the target-side low-resource Chinese dialects, and fine-tune another target ASR model based on the previous source-side ASR model. Meanwhile, the potential common semantic features between source-side and target-side ASR models can be captured by using self-attention mechanism. Finally, we extract the hidden semantic representation in the target ASR model to conduct Chinese dialects discrimination. Our extensive experimental results demonstrate that our model significantly outperforms state-of-the-art methods on two benchmark Chinese dialects corpora.

The generation of stylish Chinese fonts is an important problem involved in many applications. Most of existing generation methods are based on the deep generative models, particularly, the generative adversarial networks (GAN) based models. However, these deep generative models may suffer from the mode collapse issue, which significantly degrades the diversity and quality of generated results. In this paper, we introduce a one-bit stroke encoding to capture the key mode information of Chinese characters and then incorporate it into CycleGAN, a popular deep generative model for Chinese font generation. As a result we propose an efficient method called StrokeGAN, mainly motivated by the observation that the stroke encoding contains amount of mode information of Chinese characters. In order to reconstruct the one-bit stroke encoding of the associated generated characters, we introduce a stroke-encoding reconstruction loss imposed on the discriminator. Equipped with such one-bit stroke encoding and stroke-encoding reconstruction loss, the mode collapse issue of CycleGAN can be significantly alleviated, with an improved preservation of strokes and diversity of generated characters. The effectiveness of StrokeGAN is demonstrated by a series of generation tasks over nine datasets with different fonts. The numerical results demonstrate that StrokeGAN generally outperforms the state-of-the-art methods in terms of content and recognition accuracies, as well as certain stroke error, and also generates more realistic characters.

To enhance the effectiveness of differential evolution (DE) algorithm, in recent years, a number of DE variants have been proposed by employing the idea of multiple strategies. Although these DE variants have been shown competitive performance, there still exists a problem for them that the strategy selection mechanism mainly relies on the historical search experience. Unfortunately, the historical search experience may be suitable for the problems with a flat fitness landscape, while not for the problems with a rugged fitness landscape. To alleviate the issue, in this work, an individual-dependent DE variant, called IPDE, is proposed by using the dual information guidance, including the fitness information and spatial information. In the IPDE, the idea of multiple strategies is used as well, but its most salient feature lies in that the strategy selection mechanism is based on the individual role rather than the historical search experience. To better identify the individual role, the fitness information and spatial information are used simultaneously, which can roughly estimate the evolutionary statuses of different individuals in the fitness landscape. To validate the effectiveness of IPDE, 52 benchmark functions are used in the experiments, and four well-established evolutionary algorithms are included in the performance comparison. The final results show that IPDE can achieve promising performance.

Due to the rise of e‐commerce platforms, online shopping has become a trend. However, the current mainstream retrieval methods are still limited to using text or exemplar images as input. For huge commodity databases, it remains a long‐standing unsolved problem for users to find the interested products quickly. Different from the traditional text‐based and exemplar‐based image retrieval techniques, sketch‐based image retrieval (SBIR) provides a more intuitive and natural way for users to specify their search need. Due to the large cross‐domain discrepancy between the free‐hand sketch and fashion images, retrieving fashion images by sketches is a significantly challenging task. In this work, we propose a new algorithm for sketch‐based fashion image retrieval based on cross‐domain transformation. In our approach, the sketch and photo are first transformed into the same domain. Then, the sketch domain similarity and the photo domain similarity are calculated, respectively, and fused to improve the retrieval accuracy of fashion images. Moreover, the existing fashion image datasets mostly contain photos only and rarely contain the sketch‐photo pairs. Thus, we contribute a fine‐grained sketch‐based fashion image retrieval dataset, which includes 36,074 sketch‐photo pairs. Specifically, when retrieving on our Fashion Image dataset, the accuracy of our model ranks the correct match at the top‐1 which is 96.6%, 92.1%, 91.0%, and 90.5% for clothes, pants, skirts, and shoes, respectively. Extensive experiments conducted on our dataset and two fine‐grained instance‐level datasets, i.e., QMUL‐shoes and QMUL‐chairs, show that our model has achieved a better performance than other existing methods.

Artificial bee colony (ABC) algorithm is a relatively new bio inspired optimization technique which has attracted a lot of attention for its competitive performance. However, in the basic ABC, the solution search equation performs well in exploration but poorly in exploitation, which may cause the problem of slow convergence rate. To tackle this issue, we propose a dynamic best neighbor-guided search strategy to enhance the performance of ABC. In the proposed strategy, a dynamic neighborhood with variable size is first constructed with respect to different evolution stage of the algorithm. After that, the best food source of the neighborhood is selected to guide search instead of only using the global best food source or some elite food sources, which aims to achieve a better balance between the exploration and exploitation. In addition, we design an improved global neighborhood search operator with better robustness to further enhance the performance of ABC. In the experiments, 50 different benchmark functions are used to verify our approach, including the CEC2013 benchmark test suite. The experimental results compared with other four recently well-established ABC variants show that our approach can significantly enhance the performance of ABC.

This paper focuses on single image derain, which aims to restore clear image from single rain image. Through full consideration of different frequency information preservation and the complicated interactions between rain-streaks and background, a novel end-to-end cumulative rain-density sensing network (CRDNet) is proposed for adaptive rain-streaks removal. An effective W-Net with powerful learning ability is proposed as a key component to recover rain-invariant low-frequency signals. A cumulative rain-density classifier with a novel cost-sensitive label encoding strategy is proposed as an auxiliary network to improve discriminative power of extracted high-frequency rain-streaks through multi-task training. The proposed CRDNet has been compared with state-of-the-art methods on two public datasets. The quantitative and visual experimental results demonstrate that it can achieve excellent performance with great improvement. Related source code and models are available on github https://github.com/peylnog/CRDNet.

Language discrimination among similar languages, varieties, and dialects is a challenging natural language processing task. The traditional text-driven focus leads to poor results. In this article, we explore the effectiveness of speech-driven features toward language discrimination among Chinese dialects. First, we systematically explore the appropriateness of speech-driven MFCC features toward CNN-based language discrimination. Then, we design an end-to-end speech recognition model based on HMM-DNN to predict Chinese dialect words. We adopt attention mechanism to extract the discriminative words related to different Chinese dialects. Finally, through a CNN, we combine the word-level embedding and the MFCC-based features. Evaluation of two benchmark Chinese dialect corpora shows the appropriateness and effectiveness of the proposed speech-driven approach to fine-grained Chinese dialect discrimination compared to the state-of-the-art methods.

As a relatively new paradigm of evolutionary algorithms, artificial bee colony (ABC) algorithm has shown attractive performance in solving optimization problems. However, for some complex optimization problems, its performance is still not satisfactory. To address this concerning issue, in this paper, we designed an improved ABC variant, called DLABC, based on the division of labor theory. In the DLABC, both of the employed bee phase and the onlooker bee phase are modified by introducing new solution search equations. For the modified employed bee phase, two new control parameters are introduced in the new solution search equation, which aims to control the frequency of perturbation and the magnitude of perturbation, respectively. For the modified onlooker bee phase, a novel depth-first search framework is used which tends to allocate more computing resources to elite solutions for accelerating convergence rate. By combing these two modified phases, we attempt to balance the exploration and exploitation capabilities of ABC. Numerical experiments are conducted on the CEC 2019 100-digit challenge benchmark suite, and the DLABC is compared with the basic ABC and a recently well-established ABC variant (DFSABC_elite). The total scores of these included algorithms have shown that our proposed DLABC algorithm performed best.

In this paper, we focus on automatically colorizing single grayscale image without manual interventions. Most of existing methods tried to accurately restore unknown ground-truth colors and require paired training data for model optimization. However, the ideal restoration objective and strict training constraints limited their performance. Inspired by CycleGAN, we formulate the process of colorization as image-to-image translation and propose an effective color-CycleGAN solution. High-level semantic identity loss and low-level color loss are additionally suggested for model optimization. Our method allows using unpaired images for training and direct prediction in rgb color space, which makes training data collection much easier and more general. We train our model on randomly selected PASCAL VOC 2007 images. All ablation study on loss function and comparisons with state-of-the-art methods are performed on grayscale SUN data. The experiment results show that our improvements on training loss could achieve better content consistence and generate better reasonable colors with less artifacts. Moreover, due to the bidirectional nature of our model, our proposed method provides a by-product that gives an excellent alternative way on color image graying.

As a relatively new paradigm of bio-inspired optimization techniques, artificial bee colony (ABC) algorithm has attracted much attention for its simplicity and effectiveness. However, the performance of ABC is not satisfactory when solving some complex optimization problems. To improve its performance, we propose a novel ABC variant by designing a dynamic multi-population scheme (DMPS). In DMPS, the population is divided into several subpopulations, and the size of subpopulation is adjusted dynamically by checking the quality of the global best solution. Moreover, we design two novel solution search equations to maximize the effectiveness of DMPS, in which the local best solution of each subpopulation and the global best solution of the whole population are utilized simultaneously. In the experiments, 32 widely used benchmark functions are used, and four well-established ABC variants are involved in the comparison. The comparative results show that our approach performs better on the majority of benchmark functions.

With the development of social media and the popularity of mobile devices, it becomes increasingly easy to post rumors and spread rumors on social media. Widespread rumors may cause public panic and negative impact on individuals, which makes the automatic detection of rumors become necessary. Most existing methods for automatic rumor detection focus on modeling features related to contents, users and propagation patterns based on feature engineering, but few work consider the existence of graph structural information in the user behavior. In this paper, we propose a model that leverages graph convolutional networks to capture user behavior effectively for rumor detection. Our model is composed of three modules: 1) a user encoder that models users attributes and behaviors based on graph convolutional networks to obtain user representation; 2) a propagation tree encoder, which encodes the structure of the rumor propagation tree as a vector with bridging the content semantics and propagation clues; 3) an integrator that integrates the output of the above modules to identify rumors. Experimental results on two public Twitter datasets show that our model achieves much better performance than the state-of-the-art methods.

In this paper, we have proposed a static crowd scene analysis network via multi-branch dilated convolution block, called MDBNet. It focuses on a joint task of estimating crowd count and high-quality density map from static single image. The proposed MDBNet follows one-stage object detection framework, and consists of two parts: pre-trained convolutional layers as the front end for high-level feature extraction and cascaded multi-branch dilated convolution block as the back end for context information aggregation on different ranges. Pixel-wise objectness probabilities are predicted and regressed to generate density map. The proposed MDBNet is an easy training model with strong learning ability. We have tested it on two public datasets (ShanghaiTech dataset and the UFC_CC_50 dataset). On almost all evaluation criterions, the proposed method has achieved superior performance. Especially on structure quality criterions, including our newly introduced spatial adjusted mutual information measurement, the MDBNet reports a new state-of-the-art performance. The source code will be distributed depending on publication of our work.

Single image super‐resolution (SISR) has gained great attraction and progress in recent years. Since the SISR is an ill‐posed inverse problem, most researchers are concentrated on making efforts to learn effective and reasonable mapping functions from low‐resolution observation to its potential high‐resolution (HR) counterpart. In this study, the authors have proposed a deep residual refining based pseudo‐multi‐frame network for efficient SISR. A channel‐wise attention mechanism is employed for residual refinement. It can ease residual learning process through explicitly modelling non‐linear dependencies between channels by using global information embedding. Multiple potential HRs from different deconvolutional layers are further artificially learned, and then adaptively fused into final desired HR image. The authors call this strategy as pseudo‐multi‐frame SR. It could make full use of available redundant information possessed in hierarchical layers. They have evaluated the proposed network on several popular benchmark datasets. The experimental results have shown that the two highlights proposed can consistently boost final performance. The proposed network can outperform most of the state‐of‐the‐art methods with acceptable less parameters.

Automatic evaluation of machine translation is critical for the evaluation and development of machine translation systems. In this study, we propose a new model for automatic evaluation of machine translation. The proposed model combines standard n-gram precision features and sentence semantic mapping features with neural features, including neural language model probabilities and the embedding distances between translation outputs and their reference translations. We optimize the model with a representative list-wise learning to rank approach, ListMLE, in terms of human ranking assessments. The experimental results on WMT’2015 Metrics task indicated that the proposed approach yields significantly better correlations with human assessments than several state-of-the-art baseline approaches. In particular, the results confirmed that the proposed list-wise learning to rank approach is useful and powerful for optimizing automatic evaluation metrics in terms of human ranking assessments. Deep analysis also demonstrated that optimizing automatic metrics with the ListMLE approach is a reasonable method and adding the neural features can gain considerable improvements compared with the traditional features.

Visual tracking is one of the fundamental problems in computer vision. Recently, some deep-learning-based tracking algorithms have been illustrating record-breaking performances. However, due to the high complexity of neural networks, most deep trackers suffer from low tracking speed and are, thus, impractical in many real-world applications. Some recently proposed deep trackers with smaller network structure achieve high efficiency while at the cost of significant decrease in precision. In this paper, we propose to transfer the deep feature, which is learned originally for image classification to the visual tracking domain. The domain adaptation is achieved via some “grafted” auxiliary networks, which are trained by regressing the object location in tracking frames. This adaptation improves the tracking performance significantly both on accuracy and efficiency. The yielded deep tracker is real time and also illustrates the state-of-the-art accuracies in the experiment involving two well-adopted benchmarks with more than 100 test videos. Furthermore, the adaptation is also naturally used for introducing the objectness concept into visual tracking. This removes a long-standing target ambiguity in visual tracking tasks, and we illustrate the empirical superiority of the more well-defined task.

Visual tracking is a fundamental problem in computer vision. However, due to the (sometimes) ambiguous target information given at the first frame, it has also been criticized as less well-posed compared with other tasks with clearly-defined targets, such as object detection and semantic segmentation. In this paper, we try to evaluate the importance of object category in visual tracking by tracking objects with known object types. The proposed algorithm, termed Deep-Track with Objectness (DTO), naturally combines the state-of-the-art deep-learning-based detectors and trackers, which essentially share a large part of the network. In DTO, a deep tracker, which is scale-fixed and sensitive to small translations tracks the object in a relative short lifespan. A deep detector, which is scale-changeable and robust to pose or illumination changes guides the deep tracker in a longer lifespan. As the deep tracker and detector share the main part of their networks, no much extra computation is imposed while the performance gain is significant. We test the proposed algorithm on two well-accepted benchmarks and on both of them, the proposed method increases the tracking accuracies remarkably compared with state-of-the-art visual trackers.

Acquiring vocabulary is important when studying English, as it assists in listening, speaking, reading, and writing. In this paper, we develop an English webpage corpus (EWC) and create a word frequency list using web crawler technology. By comparing EWC word lists with the British National Corpus (BNC), we find that the BNC word frequency list possesses the feature of timeliness. We also explore primary school students’ English word recognition rates by comparing the word frequency lists of several corpora, including EWC, BNC, SUBTLEX-US, and Subtitle Corpus of Children’s BBC (CBBC). The results show that the word recognition rates for primary school children are relatively low in both general language and specific language register. Motivated by the experiment results, we finally propose some word-selection strategies for compiling English textbooks for Chinese primary school students.

Automatic evaluation of machine translations is an important task. Most existing evaluation metrics rely on matching the same word or letter n -grams. This strategy leads to poor results on Chinese translations because one has to rely merely on matching identical characters. In this article, we propose a new evaluation metric that allows different characters with the same or similar meaning to match. An Indirect Hidden Markov Model (IHMM) is proposed to align the Chinese translation with human references at the character level. In the model, the emission probabilities are estimated by character similarity, including character semantic similarity and character surface similarity, and transition probabilities are estimated by a heuristic distance-based distortion model. When evaluating the submitted output of English-to-Chinese translation systems in the IWSLT’08 CT-EC and NIST’08 EC tasks, the experimental results indicate that the proposed metric has a significantly better correlation with human evaluation than the state-of-the-art machine translation metrics (i.e., BLEU, Meteor Universal, and TESLA-CELAB). This study shows that it is important to allow different characters to match in the evaluation of Chinese translations and that the IHMM is a reasonable approach for the alignment of Chinese characters.

Latent Semantic Indexing (LSI) has been successfully applied to information retrieval and text classification. However, when LSI is used in classification, some important features for small classes may be ignored because of their small feature values. To solve this problem, we propose the latent semantic classification (LSC) model which extends the LSI model in the following way: the classification information of the training documents is introduced into the latent semantic structure via a second set of latent variables, so that both indexing and classification information can be taken into account during the classification process. Our experiments on Reuters show that our new model performs better than the existing classification methods such as kNN and SVM.